Understanding developmental bias using single cell technology

The phenotype of an adult animal has two histories. One is its evolutionary history, that is how it has evolved over hundreds of millions of years. The other is its developmental history, as with each generation it is built anew through the process of embryonic development. We already know that evolution changes the developmental process to effect phenotypic change. Much less well understood is whether the developmental process imposes restrictions on what evolution can do. Developmental mechanisms are very complex, involving thousands of genes and hundreds of thousands of regulatory elements (bits of the genome that control when and where a gene is expressed). Does this complexity impose limitations on evolution? For example, are certain types of evolutionary change more or less likely to come about through changes in specific parts of the developmental mechanism? Does this mean that there are potential phenotypes that are difficult or even impossible for evolution to access? These are known as developmental biases and uncovering them is important to understand not only our own evolutionary history, but also what we can easily change through genetic engineering or what phenotypes can be designed through synthetic biology. Up until very recently it has been extremely hard to investigate this as it requires deep, system wide understanding of development in multiple species, work that has historically taken a great many people many years to undertake for even one species. The arrival of single cell technology is changing that, as it allows system wide data to be gathered quickly and efficiently by a small team. In this project we will use this technology to collect such data on a developing system, the spinal cord of vertebrates. We have chosen this as it is one of the best understood developmental systems, and it comes with a wealth of pre-existing knowledge about mechanisms, especially about the signalling between cells that plays a critical role in its development. We will apply this technique across developmentally manipulated vertebrate species chosen for their evolutionary relationships. Galliform birds (chicken, quail and allies) allow comparison of very similar species and recent changes, while extending this approach to vertebrates from other Classes (mouse, as a biomedical model system, and lamprey as the earliest diverging vertebrate) will allow us to also look far back in time. Our specific aims are to: Generate single cell transcriptome and regulome data across the relevant species to identify where evolution has modified this developmental system. Manipulate the development of the tissues in a defined comparative framework to identify where the system imposes constraints on what does or doesn’t change. Integrate these datasets to identify where bias is operating Test whether outcomes are generalisable to other developmental mechanisms. Combined, the outcomes should tell us how the complexities of developmental mechanisms skew evolution, and help us to predict what can easily be changed in other species and tissues to effect robust phenotypic change.